Association between Periodontitis and Carotid Atherosclerosis: An Updated Systematic Review and Meta-Analysis

Periodontitis is an inflammatory disease that destroys the tissues around the teeth [1]. Severe periodontitis has been identified as the main cause of tooth loss [1, 2]. Increasing evidence suggests a close link between periodontal disease and cardiovascular disease, such as diabetes, stroke, atherosclerosis, and coronary artery disease [3‒6]. A survey based on 8,649 nationally representative participants revealed a lower risk of periodontitis in participants with higher cardiovascular health (CVH) score (higher CVH score means better CVH) [7]. Additionally, AlSakr et al. [8] reported a significant link between chronic periodontitis and intracranial carotid artery calcification.

Atherosclerosis is considered as the primary cause of cardiovascular death. The earliest macroscopic changes in vessels are fatty streaks [9]. As the disease progresses, these evolve into atheromatous plaques, resulting in occlusion of the vessel, and may precipitate a stroke or myocardial infarction [9, 10]. It has been identified that age, hypertension, diabetes, smoking, obesity, and dyslipidemia are contributing factors in the development of atherosclerosis [10]. Periodontal disease shares the same risk factors such as smoking and diabetes [11]. Periodontitis can trigger an inflammatory cascade in the body, and the host immune response also favors atheroma formation, maturation, and exacerbation [12]. Therefore, it is meaningful to identify the role of periodontitis in atherosclerotic processes.

Many studies have explored the connection between periodontitis and carotid atherosclerosis (CAS); however, conflicting findings were reported. Some studies have revealed a significant association between periodontitis and CAS [8, 13‒15], whereas others showed a nonsignificant association [9, 16]. A meta-analysis conducted by Zeng et al. [16] published in 2016 has studied the link between periodontal disease and CAS. Subsequently, some investigations have been performed regarding the association. Therefore, the study was conducted to integrate the present evidence on the relationship between periodontitis and CAS.

Our review conformed to the PRISMA statement guidelines and was registered in the International Prospective Register of Systematic Reviews (PROSPERO) database (CRD4202451007).

PubMed, Embase, Web of Science, and Cochrane Library were searched for studies published up to 14 December 2023 without language restrictions. We established the search strategy by adding mesh terms as well as keywords related to the topic through a discussion between authors. The search terms were (“carotid artery” OR “carotid”) AND (“calcium” OR “calcification” OR “plaque” OR “atherosclerosis” OR “atheroma”) AND (“Periodontitis” OR “periodontal disease” OR “periodont*”). We also screened the references of relevant reviews to identify additional studies.

After removal of duplicate publications, two investigators (Min Chen and Fangfei Ye) independently screened the titles and abstract, and a third investigator (Qun Zhou) checked for any difference. Next, a discussion was conducted to solve the discrimination. Inclusion criteria were as follows: (1) observational studies (case-control, cohort, or cross-sectional) that assessed association between periodontitis and CAS and relevant indicators, and (2) adjusted odds ratios (ORs) with 95% confidence interval (CIs) were reported. Exclusion criteria were as follows: (1) animal studies; (2) conference abstracts, reviews, comments, or case reports; (3) had duplicated data.

Two authors (Fangfei Ye and Min Chen) independently extracted the following data of included studies: name of first author, year of publication, country of origin, study design, sample size, gender of participants, assessment tool of CAS, periodontitis stage, and adjusted ORs with 95% CIs between periodontitis and CAS. Discrimination was resolved with a third author (Qun Zhou).

We assessed the quality of case-control studies and cohort studies using the Newcastle-Ottawa Scale (NOS), a validated scale including three domains: 4 points for selection, 2 points for comparability, and 3 points for exposure. Studies scored 0–3, 4–6, and 7–9 points were considered as low, moderate, and high quality, respectively. Cross-sectional studies were assessed using an adapted version of NOS with scores ranging from 0 to 10 points. Two authors (Fangfei Ye and Min Chen) independently rated the articles, and a discussion with a third author (Qun Zhou) was held if any discrimination. We considered studies of high quality with scores ≥7 points.

Statistical analysis was conducted on software STATA 14.0. The adjusted ORs with 95% CIs of individual articles assessing the associations between periodontitis and CAS were pooled into the meta-analysis using random-effects models due to differences in the methodology of studies. When periodontitis was grouped into more than one category by stage in a particular study, we calculated the combined OR using a method provided by Helming [17]. If sufficient information unavailable for Helming’s method, we combined the available single OR to obtain an overall OR using a fixed-effects model for the main meta-analysis.

Heterogeneity was assessed using Cochran’s Q and I² statistics. And p < 0.10 for the Q test or I² > 50% represents significant heterogeneity. Subgroup analysis was performed to explore the source of heterogeneity by study design, country of origin, assessment tool of CAS, study quality, and periodontitis stage. Also, sensitivity analysis was conducted by omitting one study at a time to assess the stability of the results and possible source of heterogeneity. We assessed publication bias by generating a funnel plot and carrying out Begg’s test, and Egger’s test. And p < 0.05 indicated statistical significance.

As shown in Figure 1, a total of 652 citations were collected. After removing the 309 duplicates and initial screening of titles and abstracts, 56 records were retained for full-text reading, of which 30 studies were excluded according to eligible criteria. We eventually included 26 studies for meta-analysis [8, 9, 13‒15, 18‒38].

Of the 26 studies, 19 were cross-sectional studies [8, 9, 13‒15, 18‒22, 26, 27, 29, 30, 32, 34, 36‒38], 3 were cohort study [24, 28, 35], and the remaining 4 were case-control studies [13, 23, 31, 33]. The origin countries of the studies included China, Finland, Germany, India, Italy, Japan, Korea, Mongolia, Spain, Sweden, and USA that distributed in Asia, Europe, and North America.

Of the studies, carotid arterial calcification was diagnosed by panoramic radiographs in most studies [9, 15, 20, 23, 29, 34], and two studies were diagnosed by computed tomography [8, 21]. Other markers of CAS such as carotid intima-media thickness and carotid plaque were assessed using B-mode ultrasound. All of these included studies used none/mild periodontitis as control, except for one study that used non-severe periodontitis as control [22]. Almost all studies evaluated periodontitis using one kind of definition of periodontitis; however, 2 studies assessed periodontitis using two methods [23, 29]. In the study of Gustafsson et al. [23], periodontitis was assessed by degree of alveolar bone loss and a clinical periodontal disease index (CPDI) score. Paju et al. [29] evaluated the periodontal status using alveolar bone loss or periodontal inflammation (Table 1).

The NOS scores are shown in Table 1. Three case-control studies scored ≥7 and were rated as high quality [13, 23, 33], and 1 scored 5 points was judged as moderate quality [31]. One cohort study included was classified as high quality [35], and 2 were moderate quality [24, 28]. Concerning cross-sectional studies, 1 study scored 6 points was considered as medium quality [30], and the other 18 studies were of high quality (range 8–9 points) [8, 9, 13‒15, 18‒22, 26, 27, 29, 32, 34, 36‒38]. The detailed scores are presented in online supplementary Table S1 (for all online suppl. material, see https://doi.org/10.1159/000543955).

The combined results of the 26 studies revealed a significantly positive relationship between periodontitis and CAS (OR = 1.97, 95% CI = 1.64–2.36; p = 0.000), with high heterogeneity (p = 0.000 and I² = 85.3%) (shown in Fig. 2). The outcome of sensitivity analysis showed that the pooled effect was not affected by any single study, indicating that the result is robust (shown in online suppl. Fig. S1). Therefore, we conducted a subgroup analysis according to study location, assessment tool of CAS, study quality, and periodontitis stage to explore the potential sources of the heterogeneity. We found that periodontitis was significantly associated with CAS in all subgroups with the exception of cohort study design (however, without statistical significance, z = 1.57, p = 0.117). Heterogeneity was observed in all subgroups except for mild and moderate periodontitis (Table 2).

The funnel plot was visually asymmetric, which was consistent with the result of Begg’s test (p = 0.000) and Egger’s test (p = 0.000) (shown in online suppl. Fig. S2). To address this, we used the trim-and-fill method to correct the bias. The result revealed that at least 12 hypothetical studies were needed to make the plot symmetric (shown in online suppl. Fig. S3). The correction still showed a significant increased risk of CAS for periodontitis (OR: 1.30; 95% CI: 1.06–1.58; p = 0.000).

This meta-analysis explored the association between periodontitis and CAS and revealed that periodontitis was significantly associated with CAS. In addition, our analysis presents an increased OR in association between severe periodontitis and CAS.

The association between periodontal health and cardiovascular diseases is constantly being explored. A study based on 6,017 individuals showed that IMT ≥1 mm was 2.09 for severe periodontitis and 1.40 for moderate periodontitis compared with healthy control, and severe periodontitis was still associated with IMT ≥1 mm after adjusting for confounding factors [19]. Gustafsson et al. [23] found an association between periodontitis and calcified carotid artery atheromas when periodontitis was defined by a CPDI score (from clinical and radiographic assessments); however, no association was found when periodontitis was assessed radiographically (as degree of bone loss). However, a study conducted by Paju et al. [29] reported a conflicting conclusion, and their study suggested that periodontitis has significant association with carotid artery calcification when periodontitis was evaluated according to the degree of bone loss by panoramic radiograph. This discrepancy may be due to overlapped tissue structures, where some teeth cannot be measured in panoramic radiographs. What’s more, bone loss is unable to reflect the real inflammation status. For example, it may be at the inactive period with a low risk of progression.

Several previous review and meta-analyses on a similar topic have been published. A meta-analysis published in 2016 revealed that patients with periodontal disease were associated with CAS [16], which is consistent with our findings, but there are still some differences. For instance, our study exclusively included research with a clear definition of periodontitis. A recent meta-analysis conducted by Wang et al. [39] also revealed a significant association between carotid artery calcification and periodontitis. However, we found that only cross-sectional studies were included in this review, and it is a pity that many important data were lost. Therefore, our updated meta-analysis would enhance the statistical power on the association between periodontitis and CAS.

Periodontitis may exacerbate the pathogenesis of cardiovascular disease [11, 40, 41]. At present, four viewpoints have been put forward to explain the biological mechanism by which periodontal disease exacerbates the progression of atherosclerosis that involves invasion of bacteria, autoimmune responses, and cytokine theory [42]. Some periodontal pathogens have been detected in human cardiovascular system including heart valves, pericardial fluids, cardiac tissue, and atherosclerotic lesions [43, 44]. In addition, periodontal pathogens can damage the epithelium of the periodontal pocket, thus releasing noxious endotoxins and exotoxins into the bloodstream [45]. The entry of bacteria would further activate inflammatory response by multiple mechanisms. In addition, bacteria in the plaque can trigger the host epithelial cells to release pro-inflammatory cytokines including IFN-γ, IL-1β, TNFα, and PGE2 [46‒48]. Elevating levels of some inflammatory biomarkers, such as CRP and IL-6, have been repeatedly observed in patients with periodontitis, which are considered as independent risk factors of cardiovascular disease [49]. Interestingly, several clinical studies have reported that serum levels of IL-6 and CRP were markedly decreased after subgingival nonsurgical periodontal therapy in patients with both periodontal and cardiovascular diseases in a certain period of time [50‒53].

This meta-analysis has a few limitations. First, it is unable to identify a causal relationship between periodontitis and CAS because the design of included studies was observational; long-term randomized controlled trials are thus warranted. Second, though we conducted a search on four major databases, we might miss studies focused on this topic. Third, the definitions of periodontitis and CAS are not fully consistent, which may lead to selection bias. Fourth, high heterogeneity was observed when evaluating the association between periodontitis and CAS and carotid stenosis, suggesting that there is heterogeneity among the studies. However, the outcome of sensitivity analysis suggested a robust result. Other limitations include the inability to conduct subgroup analyses, sensitivity analyses, and publication bias analyses due to limitation number of included studies. Therefore, more studies on this topic should be conducted in the future.

The findings indicate that periodontitis is significantly associated with CAS. However, it is unable to establish causality due to the observational study design of included studies. Large-scale and well-designed randomized controlled trials are needed to validate the causality.

An ethics statement is not applicable because this study is based exclusively on published literature.

The authors declare no conflict of interest.

The present study was supported by Zhejiang Provincial Traditional Chinese Medicine Science and Technology Project (2024ZR186).

Conceptualization and original drafting, F.Y.; methodological design, data collection, and calculation, F.Y., M.C., and Q.Z.; revision, M.C. All the authors have read and agreed to the published version of the manuscript.

All data generated or analyzed during this study are included in this article and its supplementary material files. Further inquiries can be directed to the corresponding author.